UAT: `agents plan prompt` JSON/YAML output missing `timing.started` field required by spec #6321

New issue

Open

opened 2026-04-09 20:10:05 +00:00 by HAL9000 · 0 comments

HAL9000 commented

2026-04-09 20:10:05 +00:00

Owner

Bug Report

Feature Area: Plan Lifecycle
Spec Reference: docs/specification.md §agents plan prompt (around line 15987)
Instance: uat-worker-W1-cycle3

Summary

agents plan prompt --format json produces a timing object that only contains duration_ms, missing the spec-required started (ISO 8601 timestamp) field.

Code Location

/app/src/cleveragents/cli/commands/plan.py, lines 3396–3403:

envelope: dict[str, object] = {
    "command": "plan prompt",
    "status": "ok",
    "exit_code": 0,
    "data": prompt_data,
    "timing": {"duration_ms": elapsed_ms},   # ← missing "started"
    "messages": ["Guidance queued"],
}

Steps to Reproduce

agents plan prompt <PLAN_ID> "Use mocks for database tests" --format json

Expected (per spec §agents plan prompt)

{
  "command": "plan prompt",
  "status": "ok",
  "exit_code": 0,
  "data": { ... },
  "timing": { "started": "2025-06-15T10:35:00Z", "duration_ms": 620 },
  "messages": ["Guidance queued"]
}

Actual

{
  "command": "plan prompt",
  "status": "ok",
  "exit_code": 0,
  "data": { ... },
  "timing": { "duration_ms": 620 },
  "messages": ["Guidance queued"]
}

Note: timing.started is absent.

Impact

API consumers relying on timing.started for latency tracing will receive undefined instead of a timestamp
Inconsistent with other plan commands that include timing.started (e.g., plan execute)
Breaks any client code that validates the timing object schema

Automated by CleverAgents Bot
Supervisor: UAT Testing | Agent: uat-tester

## Bug Report **Feature Area**: Plan Lifecycle **Spec Reference**: `docs/specification.md` §`agents plan prompt` (around line 15987) **Instance**: uat-worker-W1-cycle3 --- ### Summary `agents plan prompt --format json` produces a `timing` object that only contains `duration_ms`, missing the spec-required `started` (ISO 8601 timestamp) field. --- ### Code Location `/app/src/cleveragents/cli/commands/plan.py`, lines 3396–3403: ```python envelope: dict[str, object] = { "command": "plan prompt", "status": "ok", "exit_code": 0, "data": prompt_data, "timing": {"duration_ms": elapsed_ms}, # ← missing "started" "messages": ["Guidance queued"], } ``` --- ### Steps to Reproduce ```bash agents plan prompt <PLAN_ID> "Use mocks for database tests" --format json ``` --- ### Expected (per spec §agents plan prompt) ```json { "command": "plan prompt", "status": "ok", "exit_code": 0, "data": { ... }, "timing": { "started": "2025-06-15T10:35:00Z", "duration_ms": 620 }, "messages": ["Guidance queued"] } ``` ### Actual ```json { "command": "plan prompt", "status": "ok", "exit_code": 0, "data": { ... }, "timing": { "duration_ms": 620 }, "messages": ["Guidance queued"] } ``` Note: `timing.started` is absent. --- ### Impact - API consumers relying on `timing.started` for latency tracing will receive `undefined` instead of a timestamp - Inconsistent with other plan commands that include `timing.started` (e.g., `plan execute`) - Breaks any client code that validates the timing object schema --- **Automated by CleverAgents Bot** Supervisor: UAT Testing | Agent: uat-tester

Rows
Columns

UAT: agents plan prompt JSON/YAML output missing timing.started field required by spec #6321